Diabetes impairs left ventricular systolic and diastolic function and is associated with a number of abnormalities at the cellular level in the various steps of excitation-contraction coupling and signaling pathways. Two key abnormalities in experimental diabetic cardiomyopathy are a defect in sarcoplasmic reticulum (SR) function, which is associated with abnormal intracellular calcium handling, and a reduction in phosphatidyl-inositol 3-OH (PI3-) kinase activity and GLUT-4 expression. Deficient SR Ca2+ uptake during relaxation has been identified in diabetic hearts and has been associated with a decrease in the expression and activity of SR Ca2+-ATPase (SERCA2a). Furthermore, a decrease in PI3 kinase activity leading to a decrease in Akt activation and subsequently a decrease in GLUT-4 expression leads to impaired glucose uptake and hypertrophy. Using gene transfer to target specific pathways in cardio-myocytes, we will try to understand whether the excitation-contraction coupling changes in diabetic hearts occur before overt heart failure develops and/or participate in the initiation or the worsening of the changes. We will test the following hypotheses: 1) that the decrease in PI3-Kinase activity is linked to the EC coupling pathways and so contributes to the contractile dysfunction in diabetic hearts: 2) that metabolic interventions will protect against cardiac contractile dysfunction in diabetic hearts: and 3) that modulation of insulin-mediated signal transduction pathways in vivo can favorably modulate alterations of the EC coupling pathways and improve cardiac energetics and survival in diabetic rats. To test these hypotheses, we will use viral vectors to express wild-type and mutant forms of specific signaling molecules in cardiomyocytes in vitro and in vivo. We will examine the effects of PI3-Kinase, Akt, and GLUT-4 on contractile function, energetics and survival and remodeling in diabetic cardiomyopathic hearts. Animal models of type 1 and type 2 diabetes will be utilized in addition to primary cultures of human and adult rats ventricular myocytes.